Natools

Check-in [ab9c060396]
Login
Overview
Comment:tools/smaz: basic adaptation to use the generic framework
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: ab9c060396d68b18fa0cb1085598447059a65066
User & Date: nat on 2016-11-29 20:37:10
Other Links: manifest | tags
Context
2016-11-30
22:30
natools.gpr: add a profiling mode to the build check-in: 2a24860505 user: nat tags: trunk
2016-11-29
20:37
tools/smaz: basic adaptation to use the generic framework check-in: ab9c060396 user: nat tags: trunk
2016-11-28
22:31
smaz_tools-gnat: duplication of GNAT tools for generic-based tree check-in: 29b53deb6d user: nat tags: trunk
Changes

Modified tools/smaz.adb from [c51ed45f36] to [b5bd7e07f7].

25
26
27
28
29
30
31
32
33




34
35
36
37
38
39
40
41

42


43

44
45
46
47
48
49
50
25
26
27
28
29
30
31


32
33
34
35
36
37
38
39
40
41
42

43
44
45
46

47
48
49
50
51
52
53
54







-
-
+
+
+
+







-
+

+
+
-
+







with Ada.Strings.Fixed;
with Ada.Strings.Unbounded;
with Ada.Text_IO.Text_Streams;
with Natools.Getopt_Long;
with Natools.Parallelism;
with Natools.S_Expressions.Parsers;
with Natools.S_Expressions.Printers;
with Natools.Smaz.Tools;
with Natools.Smaz.Tools.GNAT;
with Natools.Smaz_256;
with Natools.Smaz_Generic.Tools;
with Natools.Smaz_Tools;
with Natools.Smaz_Tools.GNAT;
with Natools.String_Escapes;

procedure Smaz is
   function To_SEA (S : String) return Ada.Streams.Stream_Element_Array
     renames Natools.S_Expressions.To_Atom;

   package Holders is new Ada.Containers.Indefinite_Holders
     (Natools.Smaz.Dictionary, Natools.Smaz."=");
     (Natools.Smaz_256.Dictionary, Natools.Smaz_256."=");

   package Tools_256 is new Natools.Smaz_256.Tools;

   package Methods renames Natools.Smaz.Tools.Methods;
   package Methods renames Natools.Smaz_Tools.Methods;

   package Actions is
      type Enum is
        (Nothing,
         Decode,
         Encode,
         Evaluate);
97
98
99
100
101
102
103
104

105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126


127
128

129
130
131
132
133
134
135
136
137
138
139
140



141
142
143
144
145
146
147
148
149
150



151
152
153

154
155
156
157
158
159
160


161
162

163
164
165
166
167
168

169
170
171
172

173
174
175
176
177
178
179
180
181
182
183
184


185
186
187
188
189
190
191
101
102
103
104
105
106
107

108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128


129
130
131

132
133
134
135
136
137
138
139
140
141



142
143
144
145
146
147
148
149
150
151



152
153
154
155
156

157
158
159
160
161
162


163
164
165

166
167
168
169
170
171

172
173
174
175

176
177
178
179
180
181
182
183
184
185
186


187
188
189
190
191
192
193
194
195







-
+




















-
-
+
+

-
+









-
-
-
+
+
+







-
-
-
+
+
+


-
+





-
-
+
+

-
+





-
+



-
+










-
-
+
+







      Max_Sub_Size : Positive := 3;
      Max_Word_Size : Positive := 10;
      Dict_Size : Positive := 254;
      Vlen_Verbatim : Boolean := True;
      Max_Pending : Ada.Containers.Count_Type
        := Ada.Containers.Count_Type'Last;
      Job_Count : Natural := 0;
      Filter_Threshold : Natools.Smaz.Tools.String_Count := 0;
      Filter_Threshold : Natools.Smaz_Tools.String_Count := 0;
      Score_Method : Methods.Enum := Methods.Encoded;
      Action : Actions.Enum := Actions.Nothing;
      Ada_Dictionary : Ada.Strings.Unbounded.Unbounded_String;
      Hash_Package : Ada.Strings.Unbounded.Unbounded_String;
      Dict_Source : Dict_Sources.Enum := Dict_Sources.S_Expression;
   end record;

   overriding procedure Option
     (Handler  : in out Callback;
      Id       : in Options.Id;
      Argument : in String);

   overriding procedure Argument
     (Handler  : in out Callback;
      Argument : in String)
     is null;


   procedure Evaluate_Dictionary
     (Job_Count : in Natural;
      Dict : in Natools.Smaz.Dictionary;
      Corpus : in Natools.Smaz.Tools.String_Lists.List;
      Dict : in Natools.Smaz_256.Dictionary;
      Corpus : in Natools.Smaz_Tools.String_Lists.List;
      Compressed_Size : out Ada.Streams.Stream_Element_Count;
      Counts : out Natools.Smaz.Tools.Dictionary_Counts);
      Counts : out Tools_256.Dictionary_Counts);
      --  Dispatch to parallel or non-parallel version of Evaluate_Dictionary
      --  depending on Job_Count.

   function Getopt_Config return Getopt.Configuration;
      --  Build the configuration object

   procedure Optimization_Round
     (Dict : in out Holders.Holder;
      Score : in out Ada.Streams.Stream_Element_Count;
      Counts : in out Natools.Smaz.Tools.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Counts : in out Tools_256.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz_Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz_Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum;
      Updated : out Boolean);
      --  Try to improve on Dict by replacing a single entry from it with
      --  one of the substring in Pending_Words.

   function Optimize_Dictionary
     (Base : in Natools.Smaz.Dictionary;
      Pending_Words : in Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
     (Base : in Natools.Smaz_256.Dictionary;
      Pending_Words : in Natools.Smaz_Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz_Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum)
     return Natools.Smaz.Dictionary;
     return Natools.Smaz_256.Dictionary;
      --  Optimize the dictionary on Input_Texts, starting with Base and
      --  adding substrings from Pending_Words.

   procedure Parallel_Evaluate_Dictionary
     (Job_Count : in Positive;
      Dict : in Natools.Smaz.Dictionary;
      Corpus : in Natools.Smaz.Tools.String_Lists.List;
      Dict : in Natools.Smaz_256.Dictionary;
      Corpus : in Natools.Smaz_Tools.String_Lists.List;
      Compressed_Size : out Ada.Streams.Stream_Element_Count;
      Counts : out Natools.Smaz.Tools.Dictionary_Counts);
      Counts : out Tools_256.Dictionary_Counts);
      --  Return the same results as Natools.Smaz.Tools.Evaluate_Dictionary,
      --  but hopefully more quickly, using Job_Count tasks.

   procedure Print_Dictionary
     (Filename : in String;
      Dictionary : in Natools.Smaz.Dictionary;
      Dictionary : in Natools.Smaz_256.Dictionary;
      Hash_Package_Name : in String := "");
   procedure Print_Dictionary
     (Output : in Ada.Text_IO.File_Type;
      Dictionary : in Natools.Smaz.Dictionary;
      Dictionary : in Natools.Smaz_256.Dictionary;
      Hash_Package_Name : in String := "");
      --  print the given dictionary in the given file

   procedure Print_Help
     (Opt : in Getopt.Configuration;
      Output : in Ada.Text_IO.File_Type);
      --  Print the help text to the given file

   function To_Dictionary
     (Handler : in Callback'Class;
      Input : in Natools.Smaz.Tools.String_Lists.List)
     return Natools.Smaz.Dictionary;
      Input : in Natools.Smaz_Tools.String_Lists.List)
     return Natools.Smaz_256.Dictionary;
      --  Convert the input into a dictionary given the option in Handler


   overriding procedure Option
     (Handler  : in out Callback;
      Id       : in Options.Id;
      Argument : in String) is
257
258
259
260
261
262
263
264

265
266
267
268
269
270
271
261
262
263
264
265
266
267

268
269
270
271
272
273
274
275







-
+







            Handler.Max_Word_Size := Positive'Value (Argument);

         when Options.Job_Count =>
            Handler.Job_Count := Natural'Value (Argument);

         when Options.Filter_Threshold =>
            Handler.Filter_Threshold
              := Natools.Smaz.Tools.String_Count'Value (Argument);
              := Natools.Smaz_Tools.String_Count'Value (Argument);

         when Options.Score_Method =>
            Handler.Score_Method := Methods.Enum'Value (Argument);

         when Options.Max_Pending =>
            Handler.Max_Pending := Ada.Containers.Count_Type'Value (Argument);

279
280
281
282
283
284
285
286
287


288
289

290
291

292
293
294



295
296
297

298
299
300
301
302
303
304
305
306



307
308
309
310
311
312
313
314

315
316
317
318
319
320
321
283
284
285
286
287
288
289


290
291
292

293
294

295
296


297
298
299
300
301

302
303
304
305
306
307
308



309
310
311
312
313
314
315
316
317
318

319
320
321
322
323
324
325
326







-
-
+
+

-
+

-
+

-
-
+
+
+


-
+






-
-
-
+
+
+







-
+







            Handler.Vlen_Verbatim := False;
      end case;
   end Option;


   procedure Evaluate_Dictionary
     (Job_Count : in Natural;
      Dict : in Natools.Smaz.Dictionary;
      Corpus : in Natools.Smaz.Tools.String_Lists.List;
      Dict : in Natools.Smaz_256.Dictionary;
      Corpus : in Natools.Smaz_Tools.String_Lists.List;
      Compressed_Size : out Ada.Streams.Stream_Element_Count;
      Counts : out Natools.Smaz.Tools.Dictionary_Counts)
      Counts : out Tools_256.Dictionary_Counts)
   is
      Actual_Dict : Natools.Smaz.Dictionary := Dict;
      Actual_Dict : Natools.Smaz_256.Dictionary := Dict;
   begin
      Natools.Smaz.Tools.Set_Dictionary_For_Trie_Search (Actual_Dict);
      Actual_Dict.Hash := Natools.Smaz.Tools.Trie_Search'Access;
      Natools.Smaz_Tools.Set_Dictionary_For_Trie_Search
        (Tools_256.To_String_List (Actual_Dict));
      Actual_Dict.Hash := Natools.Smaz_Tools.Trie_Search'Access;

      for I in Actual_Dict.Offsets'Range loop
         if Natools.Smaz.Tools.Trie_Search (Natools.Smaz.Dict_Entry
         if Natools.Smaz_Tools.Trie_Search (Natools.Smaz_256.Dict_Entry
           (Actual_Dict, I)) /= Natural (I)
         then
            Ada.Text_IO.Put_Line
              (Ada.Text_IO.Current_Error,
               "Fail at" & Ada.Streams.Stream_Element'Image (I)
               & " -> " & Natools.String_Escapes.C_Escape_Hex
                  (Natools.Smaz.Dict_Entry (Actual_Dict, I), True)
               & " ->" & Natural'Image (Natools.Smaz.Tools.Trie_Search
                  (Natools.Smaz.Dict_Entry (Actual_Dict, I))));
                  (Natools.Smaz_256.Dict_Entry (Actual_Dict, I), True)
               & " ->" & Natural'Image (Natools.Smaz_Tools.Trie_Search
                  (Natools.Smaz_256.Dict_Entry (Actual_Dict, I))));
         end if;
      end loop;

      if Job_Count > 0 then
         Parallel_Evaluate_Dictionary (Job_Count,
            Actual_Dict, Corpus, Compressed_Size, Counts);
      else
         Natools.Smaz.Tools.Evaluate_Dictionary
         Tools_256.Evaluate_Dictionary
           (Actual_Dict, Corpus, Compressed_Size, Counts);
      end if;
   end Evaluate_Dictionary;


   function Getopt_Config return Getopt.Configuration is
      use Getopt;
350
351
352
353
354
355
356
357
358
359



360
361
362
363
364
365
366
367

368
369

370
371
372


373
374
375


376
377
378
379
380
381
382
383
384
385



386
387

388
389
390
391
392
393
394
355
356
357
358
359
360
361



362
363
364
365
366
367
368
369
370
371

372
373

374
375


376
377
378


379
380
381
382
383
384
385
386
387



388
389
390
391

392
393
394
395
396
397
398
399







-
-
-
+
+
+







-
+

-
+

-
-
+
+

-
-
+
+







-
-
-
+
+
+

-
+







      return R;
   end Getopt_Config;


   procedure Optimization_Round
     (Dict : in out Holders.Holder;
      Score : in out Ada.Streams.Stream_Element_Count;
      Counts : in out Natools.Smaz.Tools.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Counts : in out Tools_256.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz_Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz_Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum;
      Updated : out Boolean)
   is
      use type Ada.Streams.Stream_Element_Offset;

      New_Value : Ada.Strings.Unbounded.Unbounded_String;
      New_Position : Natools.Smaz.Tools.String_Lists.Cursor;
      New_Position : Natools.Smaz_Tools.String_Lists.Cursor;
      Worst_Index : constant Ada.Streams.Stream_Element
        := Natools.Smaz.Tools.Worst_Index (Dict.Element, Counts, Method);
        := Tools_256.Worst_Index (Dict.Element, Counts, Method);
      Worst_Value : constant String
        := Natools.Smaz.Dict_Entry (Dict.Element, Worst_Index);
      Worst_Count : constant Natools.Smaz.Tools.String_Count
        := Natools.Smaz_256.Dict_Entry (Dict.Element, Worst_Index);
      Worst_Count : constant Natools.Smaz_Tools.String_Count
        := Counts (Worst_Index);
      Base : constant Natools.Smaz.Dictionary
        := Natools.Smaz.Tools.Remove_Element (Dict.Element, Worst_Index);
      Base : constant Natools.Smaz_256.Dictionary
        := Tools_256.Remove_Element (Dict.Element, Worst_Index);
      Old_Score : constant Ada.Streams.Stream_Element_Count := Score;
   begin
      Updated := False;

      for Position in Pending_Words.Iterate loop
         declare
            Word : constant String
              := Natools.Smaz.Tools.String_Lists.Element (Position);
            New_Dict : constant Natools.Smaz.Dictionary
              := Natools.Smaz.Tools.Append_String (Base, Word);
              := Natools.Smaz_Tools.String_Lists.Element (Position);
            New_Dict : constant Natools.Smaz_256.Dictionary
              := Tools_256.Append_String (Base, Word);
            New_Score : Ada.Streams.Stream_Element_Count;
            New_Counts : Natools.Smaz.Tools.Dictionary_Counts;
            New_Counts : Tools_256.Dictionary_Counts;
         begin
            Evaluate_Dictionary
              (Job_Count, New_Dict, Input_Texts, New_Score, New_Counts);

            if New_Score < Score then
               Dict := Holders.To_Holder (New_Dict);
               Score := New_Score;
406
407
408
409
410
411
412
413

414
415
416
417
418
419
420
421
422
423
424
425
426
427
428



429
430
431

432
433
434

435
436

437
438
439
440
441
442
443
411
412
413
414
415
416
417

418
419
420
421
422
423
424
425
426
427
428
429
430



431
432
433
434
435

436
437
438

439
440

441
442
443
444
445
446
447
448







-
+












-
-
-
+
+
+


-
+


-
+

-
+








         Ada.Text_IO.Put_Line
           (Ada.Text_IO.Current_Error,
            "Removing"
            & Worst_Count'Img & "x "
            & Natools.String_Escapes.C_Escape_Hex (Worst_Value, True)
            & ", adding"
            & Counts (Dict.Element.Dict_Last)'Img & "x "
            & Counts (Dict.Element.Last_Code)'Img & "x "
            & Natools.String_Escapes.C_Escape_Hex
               (Ada.Strings.Unbounded.To_String (New_Value), True)
            & ", size"
            & Score'Img
            & " ("
            & Ada.Streams.Stream_Element_Offset'Image (Score - Old_Score)
            & ')');
      end if;
   end Optimization_Round;


   function Optimize_Dictionary
     (Base : in Natools.Smaz.Dictionary;
      Pending_Words : in Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
     (Base : in Natools.Smaz_256.Dictionary;
      Pending_Words : in Natools.Smaz_Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz_Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum)
     return Natools.Smaz.Dictionary
     return Natools.Smaz_256.Dictionary
   is
      Holder : Holders.Holder := Holders.To_Holder (Base);
      Pending : Natools.Smaz.Tools.String_Lists.List := Pending_Words;
      Pending : Natools.Smaz_Tools.String_Lists.List := Pending_Words;
      Score : Ada.Streams.Stream_Element_Count;
      Counts : Natools.Smaz.Tools.Dictionary_Counts;
      Counts : Tools_256.Dictionary_Counts;
      Running : Boolean := True;
   begin
      Evaluate_Dictionary (Job_Count, Base, Input_Texts, Score, Counts);

      while Running loop
         Optimization_Round
           (Holder,
452
453
454
455
456
457
458
459
460


461
462

463
464

465
466
467
468

469
470
471
472
473
474
475
457
458
459
460
461
462
463


464
465
466

467
468

469
470
471
472

473
474
475
476
477
478
479
480







-
-
+
+

-
+

-
+



-
+








      return Holder.Element;
   end Optimize_Dictionary;


   procedure Parallel_Evaluate_Dictionary
     (Job_Count : in Positive;
      Dict : in Natools.Smaz.Dictionary;
      Corpus : in Natools.Smaz.Tools.String_Lists.List;
      Dict : in Natools.Smaz_256.Dictionary;
      Corpus : in Natools.Smaz_Tools.String_Lists.List;
      Compressed_Size : out Ada.Streams.Stream_Element_Count;
      Counts : out Natools.Smaz.Tools.Dictionary_Counts)
      Counts : out Tools_256.Dictionary_Counts)
   is
      package String_Lists renames Natools.Smaz.Tools.String_Lists;
      package String_Lists renames Natools.Smaz_Tools.String_Lists;

      type Result_Values is record
         Compressed_Size : Ada.Streams.Stream_Element_Count;
         Counts : Natools.Smaz.Tools.Dictionary_Counts;
         Counts : Tools_256.Dictionary_Counts;
      end record;

      procedure Initialize (Result : in out Result_Values);

      procedure Get_Next_Job
        (Global : in out String_Lists.Cursor;
         Job : out String_Lists.Cursor;
504
505
506
507
508
509
510
511

512
513
514
515
516
517
518
519
520
521
522
523
524
525

526
527
528
529
530
531
532
509
510
511
512
513
514
515

516
517
518
519
520
521
522
523
524
525
526
527
528
529

530
531
532
533
534
535
536
537







-
+













-
+







      end Get_Next_Job;


      procedure Do_Job
        (Result : in out Result_Values;
         Job : in String_Lists.Cursor) is
      begin
         Natools.Smaz.Tools.Evaluate_Dictionary_Partial
         Tools_256.Evaluate_Dictionary_Partial
           (Dict,
            String_Lists.Element (Job),
            Result.Compressed_Size,
            Result.Counts);
      end Do_Job;


      procedure Gather_Result
        (Global : in out String_Lists.Cursor;
         Partial : in Result_Values)
      is
         pragma Unreferenced (Global);
         use type Ada.Streams.Stream_Element_Count;
         use type Natools.Smaz.Tools.String_Count;
         use type Natools.Smaz_Tools.String_Count;
      begin
         Compressed_Size := Compressed_Size + Partial.Compressed_Size;

         for I in Counts'Range loop
            Counts (I) := Counts (I) + Partial.Counts (I);
         end loop;
      end Gather_Result;
542
543
544
545
546
547
548
549

550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569

570
571
572
573
574
575
576
577
578
579
580

581
582
583
584
585
586
587
547
548
549
550
551
552
553

554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573

574
575
576
577
578
579
580
581
582
583
584

585
586
587
588
589
590
591
592







-
+



















-
+










-
+







      Counts := (others => 0);
      Parallel_Run (Cursor, Job_Count);
   end Parallel_Evaluate_Dictionary;


   procedure Print_Dictionary
     (Filename : in String;
      Dictionary : in Natools.Smaz.Dictionary;
      Dictionary : in Natools.Smaz_256.Dictionary;
      Hash_Package_Name : in String := "") is
   begin
      if Filename = "-" then
         Print_Dictionary
           (Ada.Text_IO.Current_Output, Dictionary, Hash_Package_Name);
      elsif Filename'Length > 0 then
         declare
            File : Ada.Text_IO.File_Type;
         begin
            Ada.Text_IO.Create (File, Name => Filename);
            Print_Dictionary (File, Dictionary, Hash_Package_Name);
            Ada.Text_IO.Close (File);
         end;
      end if;
   end Print_Dictionary;


   procedure Print_Dictionary
     (Output : in Ada.Text_IO.File_Type;
      Dictionary : in Natools.Smaz.Dictionary;
      Dictionary : in Natools.Smaz_256.Dictionary;
      Hash_Package_Name : in String := "")
   is
      procedure Put_Line (Line : in String);

      procedure Put_Line (Line : in String) is
      begin
         Ada.Text_IO.Put_Line (Output, Line);
      end Put_Line;

      procedure Print_Dictionary_In_Ada is
        new Natools.Smaz.Tools.Print_Dictionary_In_Ada (Put_Line);
        new Tools_256.Print_Dictionary_In_Ada (Put_Line);
   begin
      if Hash_Package_Name'Length > 0 then
         Print_Dictionary_In_Ada
           (Dictionary,
            Hash_Image => Hash_Package_Name & ".Hash'Access");
      else
         Print_Dictionary_In_Ada (Dictionary);
734
735
736
737
738
739
740
741
742


743
744

745
746
747
748
749
750
751

752
753
754
755

756
757
758

759
760
761
762

763
764
765
766
767
768
769

770
771
772
773
774
775

776
777

778
779
780
781
782
783
784
785
786

787
788
789
790
791
792
793
794
795


796
797
798
799
800
801
802
803
804
805
806

807
808
809
810
811
812
813
739
740
741
742
743
744
745


746
747
748

749
750
751
752
753



754
755
756
757

758
759
760

761
762
763
764

765
766
767
768
769
770
771

772
773
774
775
776
777

778
779

780
781
782
783
784
785
786
787
788

789
790
791
792
793
794
795
796


797
798
799
800
801
802
803
804
805
806
807
808

809
810
811
812
813
814
815
816







-
-
+
+

-
+




-
-
-
+



-
+


-
+



-
+






-
+





-
+

-
+








-
+







-
-
+
+










-
+







         end case;
      end loop;
   end Print_Help;


   function To_Dictionary
     (Handler : in Callback'Class;
      Input : in Natools.Smaz.Tools.String_Lists.List)
     return Natools.Smaz.Dictionary
      Input : in Natools.Smaz_Tools.String_Lists.List)
     return Natools.Smaz_256.Dictionary
   is
      use type Natools.Smaz.Tools.String_Count;
      use type Natools.Smaz_Tools.String_Count;
      use type Dict_Sources.Enum;
   begin
      case Handler.Dict_Source is
         when Dict_Sources.S_Expression =>
            return Natools.Smaz.Tools.To_Dictionary
              (Input,
               Handler.Vlen_Verbatim);
            return Tools_256.To_Dictionary (Input, Handler.Vlen_Verbatim);

         when Dict_Sources.Text_List | Dict_Sources.Unoptimized_Text_List =>
            declare
               Counter : Natools.Smaz.Tools.Word_Counter;
               Counter : Natools.Smaz_Tools.Word_Counter;
            begin
               for S of Input loop
                  Natools.Smaz.Tools.Add_Substrings
                  Natools.Smaz_Tools.Add_Substrings
                    (Counter, S, Handler.Min_Sub_Size, Handler.Max_Sub_Size);

                  if Handler.Max_Word_Size > Handler.Max_Sub_Size then
                     Natools.Smaz.Tools.Add_Words
                     Natools.Smaz_Tools.Add_Words
                       (Counter, S,
                        Handler.Max_Sub_Size + 1, Handler.Max_Word_Size);
                  end if;
               end loop;

               if Handler.Filter_Threshold > 0 then
                  Natools.Smaz.Tools.Filter_By_Count
                  Natools.Smaz_Tools.Filter_By_Count
                    (Counter, Handler.Filter_Threshold);
               end if;

               if Handler.Dict_Source = Dict_Sources.Text_List then
                  declare
                     Selected, Pending : Natools.Smaz.Tools.String_Lists.List;
                     Selected, Pending : Natools.Smaz_Tools.String_Lists.List;
                  begin
                     Natools.Smaz.Tools.Simple_Dictionary_And_Pending
                     Natools.Smaz_Tools.Simple_Dictionary_And_Pending
                       (Counter,
                        Handler.Dict_Size,
                        Selected,
                        Pending,
                        Handler.Score_Method,
                        Handler.Max_Pending);

                     return Optimize_Dictionary
                       (Natools.Smaz.Tools.To_Dictionary
                       (Tools_256.To_Dictionary
                          (Selected, Handler.Vlen_Verbatim),
                        Pending,
                        Input,
                        Handler.Job_Count,
                        Handler.Score_Method);
                  end;
               else
                  return Natools.Smaz.Tools.To_Dictionary
                    (Natools.Smaz.Tools.Simple_Dictionary
                  return Tools_256.To_Dictionary
                    (Natools.Smaz_Tools.Simple_Dictionary
                       (Counter, Handler.Dict_Size, Handler.Score_Method),
                     Handler.Vlen_Verbatim);
               end if;
            end;
      end case;
   end To_Dictionary;


   Opt_Config : constant Getopt.Configuration := Getopt_Config;
   Handler : Callback;
   Input_List, Input_Data : Natools.Smaz.Tools.String_Lists.List;
   Input_List, Input_Data : Natools.Smaz_Tools.String_Lists.List;
begin
   Process_Command_Line :
   begin
      Opt_Config.Process (Handler);
   exception
      when Getopt.Option_Error =>
         Print_Help (Opt_Config, Ada.Text_IO.Current_Error);
832
833
834
835
836
837
838
839

840
841
842
843

844
845
846
847
848
849
850

851
852
853
854
855
856
857
858
859
860


861
862
863
864
865
866
867

868
869
870
871
872
873
874

875
876
877
878
879
880
881
882
883
884
885
886
887

888
889
890
891
892
893
894
835
836
837
838
839
840
841

842
843
844
845

846
847
848
849
850
851
852

853
854
855
856
857
858
859
860
861


862
863
864
865
866
867
868
869

870
871
872
873
874
875
876

877
878
879
880
881
882
883
884
885
886
887
888
889

890
891
892
893
894
895
896
897







-
+



-
+






-
+








-
-
+
+






-
+






-
+












-
+







      use type Actions.Enum;

      Input : constant access Ada.Streams.Root_Stream_Type'Class
        := Ada.Text_IO.Text_Streams.Stream (Ada.Text_IO.Current_Input);
      Parser : Natools.S_Expressions.Parsers.Stream_Parser (Input);
   begin
      Parser.Next;
      Natools.Smaz.Tools.Read_List (Input_List, Parser);
      Natools.Smaz_Tools.Read_List (Input_List, Parser);

      if Handler.Action /= Actions.Nothing then
         Parser.Next;
         Natools.Smaz.Tools.Read_List (Input_Data, Parser);
         Natools.Smaz_Tools.Read_List (Input_Data, Parser);
      end if;
   end Read_Input_List;


   Build_Dictionary :
   declare
      Dictionary : Natools.Smaz.Dictionary
      Dictionary : Natools.Smaz_256.Dictionary
        := To_Dictionary (Handler, Input_List);
      Sx_Output : Natools.S_Expressions.Printers.Canonical
        (Ada.Text_IO.Text_Streams.Stream (Ada.Text_IO.Current_Output));
      Ada_Dictionary : constant String
        := Ada.Strings.Unbounded.To_String (Handler.Ada_Dictionary);
      Hash_Package : constant String
        := Ada.Strings.Unbounded.To_String (Handler.Hash_Package);
   begin
      Dictionary.Hash := Natools.Smaz.Tools.Linear_Search'Access;
      Natools.Smaz.Tools.List_For_Linear_Search := Input_List;
      Dictionary.Hash := Natools.Smaz_Tools.Linear_Search'Access;
      Natools.Smaz_Tools.List_For_Linear_Search := Input_List;

      if Ada_Dictionary'Length > 0 then
         Print_Dictionary (Ada_Dictionary, Dictionary, Hash_Package);
      end if;

      if Hash_Package'Length > 0 then
         Natools.Smaz.Tools.GNAT.Build_Perfect_Hash (Input_List, Hash_Package);
         Natools.Smaz_Tools.GNAT.Build_Perfect_Hash (Input_List, Hash_Package);
      end if;

      if Handler.Sx_Dict_Output then
         Sx_Output.Open_List;
         for I in Dictionary.Offsets'Range loop
            Sx_Output.Append_String
              (Natools.Smaz.Dict_Entry (Dictionary, I));
              (Natools.Smaz_256.Dict_Entry (Dictionary, I));
         end loop;
         Sx_Output.Close_List;
      end if;

      case Handler.Action is
         when Actions.Nothing => null;

         when Actions.Decode =>
            if Handler.Sx_Output then
               Sx_Output.Open_List;
               for S of Input_Data loop
                  Sx_Output.Append_String
                    (Natools.Smaz.Decompress (Dictionary, To_SEA (S)));
                    (Natools.Smaz_256.Decompress (Dictionary, To_SEA (S)));
               end loop;
               Sx_Output.Close_List;
            end if;

            if Handler.Stat_Output then
               declare
                  procedure Print_Line (Original, Output : Natural);
905
906
907
908
909
910
911
912

913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930

931
932
933
934
935
936
937
908
909
910
911
912
913
914

915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932

933
934
935
936
937
938
939
940







-
+

















-
+







                  Original_Total : Natural := 0;
                  Output_Total : Natural := 0;
               begin
                  for S of Input_Data loop
                     declare
                        Original_Size : constant Natural := S'Length;
                        Output_Size : constant Natural
                          := Natools.Smaz.Decompress
                          := Natools.Smaz_256.Decompress
                             (Dictionary, To_SEA (S))'Length;
                     begin
                        Print_Line (Original_Size, Output_Size);
                        Original_Total := Original_Total + Original_Size;
                        Output_Total := Output_Total + Output_Size;
                     end;
                  end loop;

                  Print_Line (Original_Total, Output_Total);
               end;
            end if;

         when Actions.Encode =>
            if Handler.Sx_Output then
               Sx_Output.Open_List;
               for S of Input_Data loop
                  Sx_Output.Append_Atom
                    (Natools.Smaz.Compress (Dictionary, S));
                    (Natools.Smaz_256.Compress (Dictionary, S));
               end loop;
               Sx_Output.Close_List;
            end if;

            if Handler.Stat_Output then
               declare
                  procedure Print_Line (Original, Output, Base64 : Natural);
953
954
955
956
957
958
959
960

961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978

979
980
981
982
983
984
985
986
987
988
989
990
991
992
993

994
995

996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007

1008
1009
1010
1011
1012
1013


1014
1015

1016
1017
1018
1019
1020
1021


1022
1023
1024


1025
1026
1027
1028
1029
1030

1031
1032
1033
1034
1035
1036

1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047


1048
1049
1050
1051
1052
1053
1054
1055
1056
1057


1058
1059

1060
1061
1062


1063
1064

1065
1066

1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084


1085
1086
1087


1088
1089

1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108

1109
1110

1111
1112

1113
1114
1115
1116
1117
1118
956
957
958
959
960
961
962

963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980

981
982
983
984
985
986
987
988
989
990
991
992
993
994
995

996
997

998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009

1010
1011
1012
1013
1014


1015
1016
1017

1018
1019
1020
1021
1022


1023
1024
1025


1026
1027
1028
1029
1030
1031
1032

1033
1034
1035
1036
1037
1038

1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049

1050
1051
1052
1053
1054
1055
1056
1057
1058
1059


1060
1061
1062

1063
1064


1065
1066
1067

1068
1069

1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086


1087
1088
1089


1090
1091
1092

1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111

1112
1113

1114
1115

1116
1117
1118
1119
1120
1121
1122







-
+

















-
+














-
+

-
+











-
+




-
-
+
+

-
+




-
-
+
+

-
-
+
+





-
+





-
+










-
+
+








-
-
+
+

-
+

-
-
+
+

-
+

-
+
















-
-
+
+

-
-
+
+

-
+


















-
+

-
+

-
+






                  Output_Total : Natural := 0;
                  Base64_Total : Natural := 0;
               begin
                  for S of Input_Data loop
                     declare
                        Original_Size : constant Natural := S'Length;
                        Output_Size : constant Natural
                          := Natools.Smaz.Compress (Dictionary, S)'Length;
                          := Natools.Smaz_256.Compress (Dictionary, S)'Length;
                        Base64_Size : constant Natural
                          := ((Output_Size + 2) / 3) * 4;
                     begin
                        Print_Line (Original_Size, Output_Size, Base64_Size);
                        Original_Total := Original_Total + Original_Size;
                        Output_Total := Output_Total + Output_Size;
                        Base64_Total := Base64_Total + Base64_Size;
                     end;
                  end loop;

                  Print_Line (Original_Total, Output_Total, Base64_Total);
               end;
            end if;

         when Actions.Evaluate =>
            declare
               Total_Size : Ada.Streams.Stream_Element_Count;
               Counts : Natools.Smaz.Tools.Dictionary_Counts;
               Counts : Tools_256.Dictionary_Counts;
            begin
               Evaluate_Dictionary (Handler.Job_Count,
                  Dictionary, Input_Data, Total_Size, Counts);

               if Handler.Sx_Output then
                  Sx_Output.Open_List;
                  Sx_Output.Append_String (Ada.Strings.Fixed.Trim
                    (Ada.Streams.Stream_Element_Count'Image (Total_Size),
                     Ada.Strings.Both));

                  for E in Dictionary.Offsets'Range loop
                     Sx_Output.Open_List;
                     Sx_Output.Append_Atom ((0 => E));
                     Sx_Output.Append_String
                       (Natools.Smaz.Dict_Entry (Dictionary, E));
                       (Natools.Smaz_256.Dict_Entry (Dictionary, E));
                     Sx_Output.Append_String (Ada.Strings.Fixed.Trim
                       (Natools.Smaz.Tools.String_Count'Image (Counts (E)),
                       (Natools.Smaz_Tools.String_Count'Image (Counts (E)),
                        Ada.Strings.Both));
                     Sx_Output.Close_List;
                  end loop;
                  Sx_Output.Close_List;
               end if;

               if Handler.Stat_Output then
                  declare
                     procedure Print
                       (Label : in String;
                        E : in Ada.Streams.Stream_Element;
                        Score : in Natools.Smaz.Tools.Score_Value);
                        Score : in Natools.Smaz_Tools.Score_Value);

                     procedure Print_Min_Max
                       (Label : in String;
                        Score : not null access function
                          (D : in Natools.Smaz.Dictionary;
                           C : in Natools.Smaz.Tools.Dictionary_Counts;
                          (D : in Natools.Smaz_256.Dictionary;
                           C : in Tools_256.Dictionary_Counts;
                           E : in Ada.Streams.Stream_Element)
                          return Natools.Smaz.Tools.Score_Value);
                          return Natools.Smaz_Tools.Score_Value);

                     procedure Print_Value
                       (Label : in String;
                        Score : not null access function
                          (D : in Natools.Smaz.Dictionary;
                           C : in Natools.Smaz.Tools.Dictionary_Counts;
                          (D : in Natools.Smaz_256.Dictionary;
                           C : in Tools_256.Dictionary_Counts;
                           E : in Ada.Streams.Stream_Element)
                          return Natools.Smaz.Tools.Score_Value;
                        Ref : in Natools.Smaz.Tools.Score_Value);
                          return Natools.Smaz_Tools.Score_Value;
                        Ref : in Natools.Smaz_Tools.Score_Value);


                     procedure Print
                       (Label : in String;
                        E : in Ada.Streams.Stream_Element;
                        Score : in Natools.Smaz.Tools.Score_Value) is
                        Score : in Natools.Smaz_Tools.Score_Value) is
                     begin
                        if Handler.Sx_Output then
                           Sx_Output.Open_List;
                           Sx_Output.Append_Atom ((0 => E));
                           Sx_Output.Append_String
                             (Natools.Smaz.Dict_Entry (Dictionary, E));
                             (Natools.Smaz_256.Dict_Entry (Dictionary, E));
                           Sx_Output.Append_String (Ada.Strings.Fixed.Trim
                             (Score'Img, Ada.Strings.Both));
                           Sx_Output.Close_List;
                        else
                           Ada.Text_IO.Put_Line
                             (Label
                              & Ada.Characters.Latin_1.HT
                              & Ada.Streams.Stream_Element'Image (E)
                              & Ada.Characters.Latin_1.HT
                              & Natools.String_Escapes.C_Escape_Hex
                                (Natools.Smaz.Dict_Entry (Dictionary, E), True)
                                (Natools.Smaz_256.Dict_Entry (Dictionary, E),
                                 True)
                              & Ada.Characters.Latin_1.HT
                              & Score'Img);
                        end if;
                     end Print;

                     procedure Print_Min_Max
                       (Label : in String;
                        Score : not null access function
                          (D : in Natools.Smaz.Dictionary;
                           C : in Natools.Smaz.Tools.Dictionary_Counts;
                          (D : in Natools.Smaz_256.Dictionary;
                           C : in Tools_256.Dictionary_Counts;
                           E : in Ada.Streams.Stream_Element)
                          return Natools.Smaz.Tools.Score_Value)
                          return Natools.Smaz_Tools.Score_Value)
                     is
                        use type Natools.Smaz.Tools.Score_Value;
                        Min_Score, Max_Score : Natools.Smaz.Tools.Score_Value
                        use type Natools.Smaz_Tools.Score_Value;
                        Min_Score, Max_Score : Natools.Smaz_Tools.Score_Value
                          := Score (Dictionary, Counts, 0);
                        S : Natools.Smaz.Tools.Score_Value;
                        S : Natools.Smaz_Tools.Score_Value;
                     begin
                        for E in 1 .. Dictionary.Dict_Last loop
                        for E in 1 .. Dictionary.Last_Code loop
                           S := Score (Dictionary, Counts, E);
                           if S < Min_Score then
                              Min_Score := S;
                           end if;
                           if S > Max_Score then
                              Max_Score := S;
                           end if;
                        end loop;

                        Print_Value ("best-" & Label, Score, Max_Score);
                        Print_Value ("worst-" & Label, Score, Min_Score);
                     end Print_Min_Max;

                     procedure Print_Value
                       (Label : in String;
                        Score : not null access function
                          (D : in Natools.Smaz.Dictionary;
                           C : in Natools.Smaz.Tools.Dictionary_Counts;
                          (D : in Natools.Smaz_256.Dictionary;
                           C : in Tools_256.Dictionary_Counts;
                           E : in Ada.Streams.Stream_Element)
                          return Natools.Smaz.Tools.Score_Value;
                        Ref : in Natools.Smaz.Tools.Score_Value)
                          return Natools.Smaz_Tools.Score_Value;
                        Ref : in Natools.Smaz_Tools.Score_Value)
                     is
                        use type Natools.Smaz.Tools.Score_Value;
                        use type Natools.Smaz_Tools.Score_Value;
                     begin
                        if Handler.Sx_Output then
                           Sx_Output.Open_List;
                           Sx_Output.Append_String (Label);
                        end if;

                        for E in Dictionary.Offsets'Range loop
                           if Score (Dictionary, Counts, E) = Ref then
                              Print (Label, E, Ref);
                           end if;
                        end loop;

                        if Handler.Sx_Output then
                           Sx_Output.Close_List;
                        end if;
                     end Print_Value;
                  begin
                     Print_Min_Max ("encoded",
                       Natools.Smaz.Tools.Score_Encoded'Access);
                       Tools_256.Score_Encoded'Access);
                     Print_Min_Max ("frequency",
                       Natools.Smaz.Tools.Score_Frequency'Access);
                       Tools_256.Score_Frequency'Access);
                     Print_Min_Max ("gain",
                       Natools.Smaz.Tools.Score_Gain'Access);
                       Tools_256.Score_Gain'Access);
                  end;
               end if;
            end;
      end case;
   end Build_Dictionary;
end Smaz;